Events:

2007-12-23 at 19:46 [xxx (lenngren)]
lise.pdc.kth.se, the login node of the Lenngren cluster, has been restarted to fix a filesystem issue.
2007-12-17 at 17:57
informational/forwarded - all afs/nada users: Maintenance work on some CSC servers will be performed on Monday December 17 starting at 5 pm. Most UNIX computers at CSC will be heavily affected during this time. Services like email and www will also be affected.
2007-12-05 at 23:42 [xxx (Hebb)]
Hebb is now back online again with newer software installed.
2007-11-27 at 15:46 [xxx (Hebb)]
Hebb will be offline due to a system upgrade next Wednesday, December 5th, starting at 10:00. It is expected to be back online again sometime that afternoon.
2007-11-22 at 14:52 [xxx (Hebb)]
The diskproblems are now solved.
2007-11-22 at 14:12 [xxx (lenngren)]
lenngren/lise: resuming node allocation. a few nodes set aside for further investigation.
2007-11-22 at 09:51 [xxx (lenngren)]
Lenngren/lise: pausing node allocation while searching for black sheep among nodes.
2007-11-17 at 13:18 [xxx (Hebb)]
gpfs: one disk-canister reporting HW problems. System is still operational but without redundancy on the affected disks.
2007-11-07 at 15:00 [xxx (Hebb)]
The whole machine is currently unavailable for running jobs due to fault determination. It will be unavailable at least until around 15:45 today.
2007-11-06 at 15:14 [xxx (Hebb)]
Half of Hebb is currently unavailable, probably due to hardware problems combined with database problems.
2007-10-29 at 15:45 [xxx (lucidor)]
blumino.pdc.kth.se (log in of the lucidor cluster) is hung. We will reboot it shortly.
2007-10-04 at 13:33 [xxx (SBC / CBR)]
The SBC AFS server valine.pdc.kth.se is now online again.
2007-10-04 at 12:34 [xxx (SBC / CBR)]
The AFS file-server valine.pdc.kth.se, hosting most of SBCs project volumes has unexpectedly crashed. Salvage is in progress and the volumes are expected to be online again at about 13:30 to 14:00.
2007-08-31 at 12:19
Forwarded info, for users with bits on CSC: Maintenance work on some CSC servers will be performed on Wednesday September 12 starting at 5 pm. Most UNIX computers at CSC will be heavily affected during this time. Services like email and www will also be affected.
2007-08-23 at 13:23 [xxx (SBC / CBR)]
Salvage complete.
2007-08-23 at 12:36 [xxx (SBC / CBR)]
the afs-server glycine.pdc.kth.se, mostly serving sbc, is salvaging its filesystems.
2007-08-23 at 12:34 [xxx (SBC / CBR)]
The SBC Cluster AFS file servers has unexpectedly gone down. We are working on restoring them.
2007-08-21 at 17:34 [xxx (HSM)]
The HSM system is now fully functional again.
2007-08-21 at 11:10 [xxx (HSM)]
The HSM system is currently unable to recall files that reside on tape due to a broken tape robot.
2007-08-15 at 12:16 [xxx (lenngren)]
The PDC HPC Summer School will be held August 20-31. It will have higher priority on Lenngren during exercises. Exercises are always held during office hours.
2007-07-26 at 00:43 [xxx (HSM)]
The HSM system is now back on-line again. There were some trouble during the upgrade, causing a longer downtime than expected.
2007-07-18 at 16:42 [xxx (HSM)]
The HSM system will be unavailable on Wednesday July 25th, starting at 13:00, due to it being upgraded. The downtime is estimated to a few hours.
2007-06-27 at 17:55
RAID box of AFS server "ruffe" did suddenly cease to serve files, therefore disrupting everything that was the slightest dependent on the AFS volumes located there (scheduler etc). Server is up again and doing internal salvaging. We have still to investigate the cause. Sorry for the disruptions
2007-06-26 at 14:13 [xxx (lenngren)]
The login node of the Lenngren cluster, lise.pdc.kth.se, will be restarted to resolve a resource scarcity issue which stops users from logging in to that node. No jobs should be affected by this.
2007-06-21 at 14:07
The PDC help desk will be closed Friday, 22 June 2007 for midsummer, and reopen again at 9:00 AM on Monday, 25 June. During the summer period 25 June through 17 August the help desk will have its normal open hours, 9:00 AM to 5:00 PM, but with many taking summer vacation, you are encouraged to call the help desk in advance of visiting PDC. E-mail, support@pdc.kth.se, and telephone, 08-790-7800, to the help desk will work as usual.
2007-06-20 at 09:42 [xxx (lucidor)]
The queue system on lucidor is now back in business. Enjoy.
2007-06-20 at 09:42 [xxx (lucidor)]
A userless job is making the queue system so confused that it cannot function. Investigation is ongoing.
2007-06-08 at 09:00
We are experiencing a dead stop since very early this morning. We currently think it's a broken port in a core switch.
2007-05-25 at 16:49
support at pdc.kth.se, as all support-queues of all other national centra will be off-line starting tomorrow 2007-05-26 at 16:00. We have been promised to get it back by Monday 2007-05-28. Our colleagues hosting it are beefing up their power grid.
2007-04-24 at 23:07 [xxx (lenngren)]
Pausing node allocation on lenngren (lise/juliana) as a precaution of possible failures. Allocation will be resumed tomorrot (2007-04-25.)
2007-04-24 at 15:18 [xxx (SBC / CBR)]
The Q-nodes of the SBC cluster are currently undergoing tests. They will be released back to the queue as soon as they finish.
2007-04-18 at 10:35
Informational: Due to excessive influx of mail over the past ten hours, mail to support at any snic-centra, nationwide, is sluggish. The response-time will vary.
2007-04-10 at 12:05 [xxx (SBC / CBR)]
The SBC queue is now running again.
2007-04-10 at 11:43 [xxx (SBC / CBR)]
The sbc queue is temporarily stopped to address a problem on the scheduler node. No jobs, enqueued or running, should be affected.
2007-03-30 at 17:20
Node allocation resumed after investigating unexpected high load on a set of file-servers.
2007-03-29 at 10:45 [xxx (SBC / CBR)]
The scheduler node of the SBC cluster has had a fatal hardware which has caused the sbc queue to stand still since approximatelly 17:00 2007-03-28. The hardware has now been replaced and the queue is running again. Please note that no jobs should have been lost/affected by this (except for a longer queue time of course).
2007-03-20 at 04:27 [xxx (lucidor)]
The log-in node (h05n35) ran out out VM (memory was overused) a couple of hours ago, and has now been restarted.
2007-03-16 at 13:46
Informational, forwarded for CSC users:Maintenance work on some CSC servers will be performed on Saturday March 24 starting at 9 am. Some work may also need to be performed on Sunday March 25, so there may be disturbances then too.
2007-03-06 at 13:37 [xxx (HSM)]
All parts of the hsm-system fully online.
2007-03-05 at 20:12 [xxx (HSM)]
the HSM is re-enabled for users. (system initiated functions i.e. through cron still pending)
2007-03-03 at 04:14 [xxx (HSM)]
the HSM system is currently inoperational depending on hardware maintenance
2007-03-01 at 06:19 [xxx (lucidor)]
Allocation resumed.
2007-02-28 at 16:35 [xxx (lucidor)]
Node allocation paused until log-in node system environment recovered. Allocation might be resumed tomorrow, 2007-03-01, at the best.
2007-02-28 at 12:13 [xxx (HSM)]
there are problems migrating data between disk and tape.
2007-02-16 at 20:02 [xxx (lucidor)]
Another Gig-E switch had to be reset. You might have experienced problems to log in.
2007-02-16 at 16:00 [xxx (lenngren)]
juliana.pdc.kth.se, the login node of the Juliana cluster will be restarted 2007-02-16 at 16:00 to clear a kernel error.
2007-02-07 at 10:15 [xxx (lucidor)]
The primary gigE etherswitch on lucidor had a fault. Most certainly all running jobs got stuck. They won't charge any accounts.
2007-02-07 at 10:12 [xxx (lucidor)]
Lucidor has lost its connection to the rest of the world. Investigation in progress.
2007-01-31 at 13:00
The main PDC web server will be relocated on the 31th of January at 13:00. We expect this to cause a short outage in web access to PDC (no other systems and or services should be affected however).
All flash news for 2018, 2017, 2016, 2015, 2014, 2013, 2012, 2011, 2010, 2009, 2008, 2007, 2006, 2005, 2004, 2003, 2002, 2001, 2000, 1999, 1998, 1997, 1996, 1995

Back to PDC
Subscribe to rss